Design Engineers - Dumb Enough to Try, Smart Enough to Pull it Off
By my own standard, I have been a successful engineer. My standard is relatively modest; I wanted to make my living by designing machines. Honestly that is all that I ever wanted from my career. I did have some bonus points along the way. I was successful enough that I was trusted with taking the lead on designs, and have had the opportunity to become a manager.
As a manager, I am in the position where I hire new engineers both directly out of college and with work experience. As a hiring manager I need to determine what I am looking for in an engineer. I spent a lot of time thinking about this, going over resumes, interviewing, and reviewing my notes from interviews. What was I looking for? What made me successful and the engineers I respected successful? What I figured out was that we were dumb enough to try and smart enough to pull it off. Designing something is hard, really, really hard. If I am honest I should just go become a bartender. Actually that is my plan when I retire. I have always wanted to be a bartender. I think it is because for some bizarre reason my dad let me watch Cheers as a kid. I always wanted to be Sam Malone.
Objectively speaking, the path from new concept to completed product is full of missing requirements, promising design schemes that turn out to not work, design compromises that seem reasonable at the time but come back to bite you, failed tests, hard lessons, assumptions that turn out to be wrong and a lot of early mornings and late nights. Not even the most seasoned design engineer sees these hurdles at the beginning. The engineer sees a shiny new problem, just waiting to be solved. If I were smarter, I would have found a different line of work. But I am not, and I fall for it every single time and dive head first into a new design.
But, and this ‘but’ is so big Sir Mix-a-Lot would approve, one cannot simply be dumb enough to try. The real world is brutal and does not care how hard an engineer tried. The flaws in the design will be exposed. Even the best designs have flaws that are exposed and must be corrected. Bad designs will never work because of the flaws. Sometimes the sheer quantity of flaws is too much for the design to survive and sometimes a flaw will be so great the only way to fix it is to start over. This is where the engineer needs to be smart enough to pull off the design.
For the non-engineer it might be concerning reading about the failed tests, hard lessons, and bad assumptions. Creating a successful design is more about understanding the problem. We typically learn more from a failed test than we do a successful test. The successful engineer is the one smart enough to learn and apply the lessons from each setback. The engineer will not look at the test as a pass or fail, but will follow up and track down exactly why a test failed. Pass or fail is not helpful terminology for an engineer. I think the best terminology is helpful or unhelpful. A test where nothing breaks is somewhat helpful, but a test when something breaks is the most helpful. When studied properly, the broken parts are the key to understanding very complex systems more fully and reveal secrets to systems that a ‘successful’ test never will. Of course, I am excluding tests that are set up incorrectly and break due to conditions the design will not experience. Those are unhelpful tests.
In modern engineering the testing takes place in both digital and real world. Finite Element Analysis (FEA) is a method engineers have developed to create digital copies of designs and test them without having to build real prototypes. In a week I could run FEA’s, that if I attempted to do in real world testing would cost millions of dollars and take years to complete. But how did we get the information required to create FEA, we ran a lot of real world tests and compiled the data in such a way that it could be used for all sorts of designs.
Then there is the small matter of time and money. If you haven’t picked up on it yet, engineers love testing. We could go on testing and learning until the end of time. There isn’t a company that has ever or will ever exist that has enough money to pay for all the testing engineers would want to do. Additionally, products do need to go to market. It isn’t always about being first to market, but that is important as well. The engineer is given a fixed budget of both time and money which requires him to be smart enough to pick the right tests that are going to tell him enough so that he can complete a safe, effective product.
There is a great middle ground of intelligence required to engineer a product. The engineer must be dumb enough to forget the late nights, ‘failed’ tests, and frustration, but smart enough to learn, adapt, choose the right tests and make the right trade offs. We aren’t the smartest people out there, but we are the ones that get it done.